High shear stresses predicted by finite element analysis in the location at which stress fractures occur in our rabbit model suggest that high shear stresses contribute to the etiology of stress fractures. We propose to test the hypotheses that (1) large bending strains cause locally high shear stress concentrations at sites where fractures occur and (2) the fatigue life of bone subject to shear stress is shorter in vivo and ex vivo. The first experiment will test the hypothesis that stress fractures occur at sites of high shear strain in the rabbit model. Rosette strain gages will be attached to the anterior cortex of the distal tibia, where stress fractures occur, and to the midshaft of 20 rabbits. The animals will be loaded repetitively with the device we used to develop the model. An FEM of each rabbit tibia will be constructed and validated against the midshaft strain state. Measured shear strains in the distal tibia will be compared to predicted shear strains at this site to determine whether the magnitude and location of predicted and measured high shear strain correspond. The second experiment will test whether stress fractures occur at the site of high shear stress if the point of high shear stress is moved from the anterior tibial cortex to the posterior cortex. This experiment has five steps: (1) Design, build and test a loading apparatus that alters the location of high shear stress; (2) define the midshaft strain state with the new apparatus; (3) construct an FEM of the rabbit tibia; (4) verify the locations of high shear strain predicted by FEM; (5) apply loads to 24 rabbits for 6 weeks to determine whether stress fractures occur at the new location of high shear stress. The third experiment will determine whether shear stress rather than normal stress dictates failure in the rabbit stress fracture model. This will be accomplished by performing shear fatigue tests. Bovine mid-diaphyses will be machined into shear test specimens. Iosipescu shear specimens and Arcan shear specimens will be used to induce pure shear in the transverse and longitudinal directions respectively. All tests will be conducted in an environmental chamber under stress control at an initial strain range of 500, 1500 and 3000 microstrain using with an MTS Bionix testing machine. If bone's fatigue life in shear is not significantly shorter than its fatigue life in uniaxial compression or tension, then the hypothesis that shear stress is the primary predisposing mechanical factor for stress fractures must be rejected. Results from this experiment will demonstrate the pathophysiological mechanisms that underlie stress fractures. If shear stresses cause stress fractures, training programs that minimize the development of shear stresses in locations at risk for stress fracture can be developed for military training and sports medicine applications.